144 Wisconsin Academy of Sciences , Arts and Letters. 
viz.: Is it conceivable that denominative verbs in simple -go should be 
formed from stems in -a (Ion., Attic -rf) e. g., a */3?}xgo from *fifixrj? Could 
we expect anything else from ^fiffxrj than ^fipxdco or *(irjxs go f Would not 
the overwhelming abundance of denominatives like rijudco, dpaojucn, 
<piXeco , etc., be a sufficient assurance that, even admitting the existence of 
fgKT], we should never have had a form ^figxco from it, from which to 
produce a perfect fie- firjx-a? 
In view of the objections which Curtius’s theory raises against itself Brug- 
mann thinks we must try in another way to solve the problem of the -xa- 
perfect. From an examination of verbal formations in -k- Brugmann 
finds six which are pre Homeric. These are: 
oXsXGO 
Sgoxoi (on the Idalian Bronze Tablet, line 16.) 
dsididdopai (for SsiSixiojuai.) 
The three aorists sdcoxa, efbrjxa, pxa. 
To all of these forms correspond perfects in -xa, which likewise seem to 
belong to the primitive stages of the language. These are 6A.GoA.sxa, Ssdooxa , 
SsiSoixa (i. e., *8sSFoixa), rsSr/xa (reSeixa), sixa. 
Brugmann subjects each of these verbs to a careful examination with 
reference to determining whether their -x- not only is pre-Homeric but pos¬ 
sibly even dates from the Ursprache. He arrives at a negative conclusion 
with reference to the forms rjxa , eSrjxa, Seididdopai, with their corres¬ 
ponding perfects sixa, rsPr/xa (re P sixa), 8 eiS oixa.) 
With reference to o’A.sxgo and its perfect oAoddexa, he finds it impossible 
to reach a decision, owing to its obscure etymology, s8coxa with its perfect 
8 sdcoxa and its present optative d goxoi is the only one of the -^-formations 
which Brugmann concedes goes back to the Ursprache for its origin. 
This form Ssdooxa Brugmann derives from the root 8 cox- and identifies 
with the Sanscrit da-ddg-a. The correspondence in form between the 
two is exact. The signification also of the Sanskrit da-ddg-a is the same 
as that of 8sdcoxa. For the Sanskrit root dag- means ‘give, vouchsafe, 
offer,’ just as does the root da- = Greek Sgo-. 
The so called aorist active sdcoxa Brugmann also derives from this same 
root 8 cox- (Sanskrit dag-). This sdcoxa is for an original ^eSoox-m, m rep¬ 
resenting the vocalic nasal or “ nasalis sonans,” which when unaccented 
regularly develops in Greek as a; just as the Rhenish Germans say gegange 
for ge-gang-n (i. e., gegangen ), geriite for gerittn(i. e., geritten), obe for obm 
(i. e., oben). Other examples in Greek of the development of the vocalic 
v ,, „ r 
nasal or “nasalis sonans” are s\v6a (for * sAivdm ), ysypacparai 
rerpdqoarai (for *yeypdcp-nrai, etc.; cf. fis-fiovXsv-rrai). 
According to the above view of the forms s8Goxa and 8s8coxa we have, 
in the different systems of the Greek verb didcopi, to deal with two roots, 
one root 8go- corresponding to the Skrt. da- and another 8cox- corres¬ 
ponding to Skrt. dac- Both roots are nearly identical in meaning and 
